Econometrics

Author: Pan Shengchu
Publisher:
Publish Date: 2002-12-01
Features: This book is revised based on the lecture notes used by the author to teach "Econometrics" to undergraduate students at the University of Finance and Economics. Since the "Econometrics" course was introduced in the late 1980s, it has been more than ten years. During this period, with the continuous development of this discipline, the lecture notes have been revised multiple times with the goals of: (1) keeping the teaching content up-to-date with the new developments in econometrics and reflecting the new achievements in research and teaching in this field; (2) making the teaching content as suitable as possible for students majoring in finance and economics, enabling them to better understand and grasp the essence of econometric theory and methods, and apply them effectively. After years of teaching practice, it can be said that satisfactory progress has been made in both aspects.
The book is divided into eight chapters: Chapter 1, Introduction; Chapter 2, Statistical Foundations of Econometrics; Chapter 3, HK Bivariate Linear Regression Model; Chapter 4, Multivariate Linear Regression Model; Chapter 5, Problems and Solutions in Model Building and Estimation; Chapter 6, Dynamic Economic Models: Autoregressive Models and Distributed Lag Models; Chapter 7, Time Series Analysis; Chapter 8, Simultaneous Equation Models.
Chapter 1 is an overview of the book, introducing what econometrics is and its history and development. It demonstrates the steps of solving problems using econometric methods with a simple example and discusses the application areas of econometrics and the software tools used.
Chapter 2 reviews the statistical concepts and methods used in econometrics, which are crucial for understanding the later content of the book. I have found that students' main difficulties often do not come from econometrics itself but from being unfamiliar with or having forgotten a large number of statistical concepts and methods. Although finance and economics majors have studied probability theory and mathematical statistics, it is unrealistic to ask them to review the entire statistics course. Even if they do so, it is often inefficient and does not address the key points. Therefore, this chapter is necessary to help students with a certain foundation in statistics quickly refresh their knowledge without having to review the entire statistics course.
Chapters 3 and 4 introduce regression analysis methods. Chapter 3 provides a detailed introduction to the concepts of the bivariate linear regression model and the least squares estimation method, as well as methods for hypothesis testing and prediction using the estimated model. Chapter 4 extends the results of the bivariate linear regression model to the multivariate linear regression model, with theoretical derivations supported by the powerful tool of matrix algebra. In the early days of econometrics teaching in Western countries, there was a tendency to avoid advanced mathematical tools as much as possible, which was related to the weak mathematical foundation of economics students. This trend has changed in recent years. My teaching practice shows that the advanced mathematics and linear algebra knowledge of Chinese economics students are sufficient to handle most derivations and proofs in regression analysis. Therefore, these two chapters provide relatively complete theoretical derivations and proofs. Of course, not all of them need to be taught in class; some can be left for interested students to review independently.
Chapter 5 discusses the problems frequently encountered in regression analysis practice and their solutions, including misspecification, multicollinearity, heteroscedasticity, and autocorrelation. Traditional methods teach these topics scattered across several chapters, but the advantage of grouping them together in this book is that it strengthens students' systematic understanding of the problems they might encounter in practice and helps them deeply grasp the connections and differences between various issues.
Chapter 6 introduces two commonly used dynamic economic models: autoregressive models and distributed lag models. The content of this chapter is arranged basically following traditional methods, with a focus on the estimation and application of these two types of models.
Chapter 7 introduces time series analysis. Time series analysis is a field in which econometrics has achieved rapid development in recent years, to the extent that many economics departments in Western universities now require courses in time series econometrics for graduate students. To keep up with this trend, it is necessary to include content on this topic in undergraduate econometrics teaching. Obviously, a single chapter is far too short to fully introduce the content of time series econometrics. Therefore, this chapter focuses on introducing some basic concepts used in time series analysis, including non-stationarity, unit roots, cointegration, and the corresponding testing methods, to give students a preliminary understanding of this field and lay a foundation for further learning and research.
Chapter 8 follows a traditional arrangement, introducing the concepts and terminology of simultaneous equation models, discussing the mathematical problems related to simultaneous equation models—identification issues—and focusing on the estimation methods of simultaneous equation models: single-equation methods and system estimation methods, as well as the most important type of simultaneous equation models—macroeconomic econometric models.
After the teaching content of each chapter, a summary is provided, which is a concise overview of the main content of the chapter. At the end of each chapter, exercises are included.